AIM: The aim of the study was to assess excess mortality related to disability pension (DP) status and DP diagnoses in Norway and Sweden during 1990-96. METHODS: Representative samples of the population aged 30-59 years, without DP at baseline 1 January 1990, 71,293 women and 76,928 men from Norway, and 68,181 women and 71,950 men from Sweden, were followed up during 1990-96. Granting of DP, DP diagnosis, age and gender were explanatory variables in Cox proportional hazards analysis with death from all causes as the outcome variable. RESULTS: Among women, 10.4% in Sweden and 7.1% in Norway obtained DP, as compared to 7.5% and 5.6% of the men. In Sweden, 66% of female and 49% of male DP recipients had musculoskeletal diagnoses, as compared to 40% and 27% in Norway. In Sweden, 3.0% of the women and 6.1% of the men with DP died, as compared to 4.6% and 8.5% in Norway. Hazard ratios (HRs) for women with DP vs. the non-DP group were 3.2 (95% confidence interval (CI)=2.7-3.8) in Sweden, and 4.9 (95% CI=4.1-5.7) in Norway. Among men with DP, there was no difference in mortality rate between the countries. HRs for men with musculoskeletal diagnoses vs. the non-DP group were 1.5 (95% CI= 1.1-2.0) in Norway and 1.4 (95% CI= 1.1-1.8) in Sweden. In both countries, the mortality rate among female disability pensioners with musculoskeletal diagnoses was not increased. CONCLUSIONS: The study confirmed an increased mortality rate among disability pensioners, except for women with musculoskeletal diagnoses. The mortality pattern related to DP diagnoses was similar in the two countries. A high frequency of musculoskeletal DP diagnoses among women with DP in Sweden explained a lower mortality rate as compared to Norway.